A system for detecting, analyzing, and resolving unmanned aerial vehicle (UAV) flight conflicts.

Abstract

This invention provides a system for detecting, analyzing, and resolving drone flight conflicts. The system includes: a data acquisition and preprocessing module for acquiring and preprocessing low-altitude basic environmental data, flight plan data, and real-time drone operation data; a drone operation risk assessment module for conducting multi-dimensional comprehensive risk assessment and analysis of the drone flight environment and finding optimal solutions to flight conflicts during drone flight; a drone operation risk alarm module for providing drone risk alarm services and drone risk solutions based on different monitoring scenarios and risk configurations; and a drone simulation module for simulating different drone flight scenarios. This invention combines pre-flight flight plan optimization with real-time conflict monitoring, achieving improved safety and management efficiency for drone flights in low-altitude environments, reducing the burden of manual monitoring, and possessing broad application prospects and practical value.

Description

Technical Field

[0001] This invention relates to the field of data digitization management system technology, and in particular to a system for detecting, analyzing and resolving unmanned aerial vehicle (UAV) flight conflicts. Background Technology

[0002] As the performance of unmanned aerial vehicles (UAVs) continues to improve and their application scope expands, their applications have become extremely broad, covering multiple fields such as military, civilian, and commercial. For example, UAVs are used for aerial photography and videography in the entertainment and film industry; for monitoring and spraying crops in agriculture; for express delivery in logistics; for aerial reconnaissance and rescue assistance in disaster relief; for monitoring, patrol, and law enforcement in urban management; and for inspection in industrial environments. UAVs are widely used in multiple business sectors, and their importance and usage are increasing daily. However, this also brings new problems. Because airspace resources are limited, the large-scale application of UAVs in the same airspace will face more flight safety issues, such as UAV collisions and intrusions into no-fly zones. Especially in dense low-altitude environments, how to effectively monitor and manage UAV operations and avoid operational risks has become an urgent problem to be solved.

[0003] Existing drone operation monitoring relies heavily on manual labor, and there is a significant lag in the timeline from the occurrence of operational risks to their discovery and eventual handling, making it impossible to detect potential risks during drone operation in advance. Furthermore, drone operation monitoring is primarily conducted from a two-dimensional or three-dimensional spatial perspective, lacking information on time and flight operations, failing to provide a comprehensive understanding of the drone's flight status, and unable to effectively address complex three-dimensional spatial and airspace environments, easily overlooking potential flight risks. Summary of the Invention

[0004] In view of this, to achieve the above-mentioned objectives, the present invention provides a system for detecting, analyzing, and resolving unmanned aerial vehicle (UAV) flight conflicts, the system comprising:

[0005] Data acquisition and preprocessing module: used to acquire and preprocess low-altitude basic environmental data, flight plan data, and real-time UAV operation data;

[0006] Drone Operation Risk Assessment Module: This module is used to analyze drone flight trajectory prediction, collision grid detection, conflict space analysis, spatiotemporal grid flight time analysis, and real-time flight conflict event analysis during drone flight to output future flight conflict events and the optimal solutions to these events.

[0007] Drone operation risk alarm module: It is used to provide drone risk alarm services and drone risk solutions based on the flight conflict events output by the system in real time, combined with different monitoring scenarios and risk configurations of the system;

[0008] The UAV simulation module is used to simulate flight plans and potential flight conflicts, and to simulate different UAV flight scenarios.

[0009] Furthermore, the data acquisition and preprocessing module includes:

[0010] Low-altitude basic environmental data acquisition unit: used to acquire geographic environmental data and airspace data to form air-ground information fusion data, and combine geospatial grid technology, system flight operation data, historical airspace flight data and future predicted airspace flight data to output multi-dimensional airspace raster;

[0011] Flight plan data acquisition unit: used to acquire UAV flight plan data, analyze and integrate the time, space and low-altitude flight business attributes of UAV flight plan data, and perform rasterization processing, that is, to perform spatiotemporal raster matching and binding on the flight routes or flight areas within the UAV flight plan to determine the associated flight spatiotemporal raster data.

[0012] The UAV real-time operation data acquisition unit is used to acquire real-time flight trajectory data, flight spatiotemporal grid data, and real-time meteorological data of the UAV. The data is filtered and fused through data fusion algorithms to form flight data features and then standardized.

[0013] Furthermore, the drone operation risk assessment module includes:

[0014] Flight trajectory prediction unit: used to predict the future flight trajectory of UAVs using multi-dimensional attribute airspace raster, flight spatiotemporal raster data and flight data features;

[0015] Collision grid detection unit: used to detect whether there are intersecting or close spatiotemporal grids in the future flight trajectory of the UAV. For spatiotemporal grids that intersect or are close to the future flight trajectory of the UAV, spatiotemporal conflict analysis is performed in combination with flight business attributes to determine whether there are spatiotemporal grids that may collide with the future flight trajectory of the UAV.

[0016] Conflict Space Analysis Unit: Used to perform spatial conflict analysis on spatiotemporal grids that have collisions;

[0017] Conflict Time Analysis Unit: Used to divide the spatiotemporal grids with collisions into conflict times;

[0018] Real-time Flight Conflict Event Analysis Unit: Used to determine future flight conflict events and find the optimal solution for such events.

[0019] Furthermore, the future flight trajectory of the drone is predicted, specifically as follows:

[0020] S11. Clean and extract features from historical flight data, and combine multi-dimensional attribute airspace raster, flight spatiotemporal raster data and flight data features to establish a Gaussian mixture prediction model for UAV flight trajectory prediction.

[0021] S12. Define the disturbance of the UAV itself and the disturbance of the airspace environment, and combine them with the Gaussian mixture prediction model to output flight trajectory prediction data.

[0022] S13. Continuously compare the predicted flight trajectory data with the actual flight trajectory data to evaluate and adjust the Gaussian mixture prediction model.

[0023] Furthermore, the spatial conflict analysis of the spatiotemporal grids where the UAVs collide specifically involves analyzing and recording potential conflict points in the three-dimensional horizontal and vertical directions of the spatiotemporal grids where collisions occur using a spatial topology analysis algorithm. This involves combining geographical location coordinates, altitude information, and three-dimensional oblique image geographic information into a three-dimensional space for analysis.

[0024] Furthermore, the determination of the future flight conflict events specifically includes:

[0025] S21. Based on the current flight status of the UAV, combined with flight plan data, real-time flight data and airspace grid data, the reachability set analysis algorithm is used to determine whether there is a conflict between the UAV and its relative position at a certain future moment.

[0026] S22. By continuously receiving real-time actual flight data and real-time airspace environment information, compare the judgment result of step S21 with the actual flight data, adjust the reachability set analysis algorithm in real time, and analyze the set of future flight trajectories that the UAV may reach within a certain time range.

[0027] S23. Use the A* algorithm to analyze all points in the future flight trajectory set, eliminate data that cannot be reached by the future flight trajectory set, and update the existing reachable dataset to determine future flight conflict events.

[0028] Furthermore, the optimal solution to the analyzed flight conflict events is sought, specifically as follows:

[0029] S31. Using flight conflict events as initial information, initialize the spatiotemporal grid, obstacles, and flight data of the conflict as information for each path;

[0030] S32. Summarize historical flight trajectory data and current airspace flight data, and decompose conflicting paths into multiple flight segments according to probability rules;

[0031] S33. Using the ant colony algorithm, find the optimal path that avoids conflict by combining multiple flight segments based on the pheromones on each conflict path.

[0032] S34. Combining the current flight speed of the UAV and airspace environment information, assign time attributes to the starting point of the flight segment on each optimal path that can avoid conflict and the path that conflicts with the original analysis.

[0033] S35. An annealing simulation algorithm is adopted, with the optimal path, flight time, spatiotemporal grid and real-time flight information as data inputs, and airspace conflict events as the stopping condition. The algorithm finds the globally optimal solution in terms of time and path through repeated iterations.

[0034] Furthermore, the drone operation risk alarm module includes:

[0035] Risk alarm unit: Used to provide drone risk alarm service. When the risk value exceeds the threshold, the system immediately sends an alarm message.

[0036] Risk resolution unit: used to provide drone risk solutions, which are used to instruct the drone to adjust at least one of flight altitude, flight speed and flight path.

[0037] Furthermore, the UAV simulation module includes:

[0038] Simulation unit: Used to combine data from the UAV operation risk assessment module and the UAV operation risk alarm module to simulate flight scenarios under different flight conditions;

[0039] Visual Interface Unit: Used to provide users with a visual operating interface to display the results and solutions of drone risk assessment and analysis.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] This invention provides a system for detecting, analyzing, and resolving UAV flight conflicts. By adding a time dimension and spatial flight business attributes, the system can more accurately locate and dynamically analyze airspace flight data, offering a more comprehensive perspective, higher precision, more efficient data processing capabilities, and better business decision support. It can also perform pre-flight environment analysis and simulation before UAV flight, as well as in-flight monitoring and real-time conflict event analysis, providing effective risk solutions. Through the automated application of simulation scenarios and conflict analysis algorithms, the system effectively improves the safety and management efficiency of UAV flights in low-altitude environments, mitigates flight conflict risks, and reduces the burden of manual monitoring, demonstrating broad application prospects and practical value. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only preferred embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of the overall structure of a system for detecting, analyzing, and resolving UAV flight conflicts, provided by an embodiment of the present invention.

[0044] Figure 2 This is a flowchart of the UAV future flight trajectory prediction process in the UAV operation risk assessment module provided in this embodiment of the invention;

[0045] Figure 3 This is a flowchart of the flight conflict event analysis in the UAV operation risk assessment module provided in this embodiment of the invention;

[0046] Figure 4 This is a flowchart illustrating the optimal solution for finding flight conflicts in the UAV operation risk assessment module provided in this embodiment of the invention. Detailed Implementation

[0047] The principles and features of the present invention are described below with reference to the accompanying drawings. The listed embodiments are only used to explain the present invention and are not intended to limit the scope of the present invention.

[0048] Reference Figure 1 This embodiment provides a system for detecting, analyzing, and resolving unmanned aerial vehicle (UAV) flight conflicts. The system includes:

[0049] Data acquisition and preprocessing module: used to acquire and preprocess low-altitude basic environmental data, flight plan data, and real-time UAV operation data.

[0050] The data acquisition and preprocessing module includes:

[0051] Low-altitude basic environmental data acquisition unit: used to acquire geographic environmental data and airspace data to form air-ground information fusion data, and combine geospatial grid technology, system flight operation data, historical airspace flight data and future predicted airspace flight data to output multi-dimensional airspace raster.

[0052] In this embodiment, the system combines acquired geographic environment data and airspace data to form air-ground information fusion data. Airspace data includes data on suitable flight zones, no-fly zones, and controlled areas. The air-ground information fusion data includes, but is not limited to, facility geographic data, oblique imagery, and airspace electronic fences. Facility geographic data and airspace electronic fence data support importing data in formats such as CSV, XML, KML, and KMZ. Oblique imagery data supports importing and loading data in formats such as OSGB, OBJ, FBX, STL, and 3Dtiles. By combining geospatial grid technology, system flight operation data, historical airspace flight data, and future predicted airspace flight data, a multi-dimensional attribute airspace raster is output. Compared to traditional multi-dimensional attribute airspace raster designs based on three-dimensional spatial element structures under low-altitude airspace restrictions, this raster includes flight operation attributes in addition to time and three-dimensional space. Flight operation attributes integrate flight volume, flight risk coefficients, plan information, noise restrictions, speed restrictions, and regional wind resistance information. The airspace is further refined into different levels of flight spatiotemporal rasterization, with each raster assigned a unique identifier.

[0053] Flight plan data acquisition unit: used to acquire UAV flight plan data, analyze and integrate the time, space and low-altitude flight business attributes of UAV flight plan data, and perform rasterization processing, that is, to perform spatiotemporal raster matching and binding on the flight routes or flight areas within the UAV flight plan to determine the associated flight spatiotemporal raster data.

[0054] In this embodiment, the unit primarily acquires and preprocesses future UAV flight plan data. The system analyzes and integrates the time, space, and low-altitude flight business attributes of the collected UAV flight plans. First, UAV flight plan data is manually entered into the system or imported into an Excel spreadsheet using a provided data template. Second, spatiotemporal grid matching and binding are performed on the flight routes or areas within the UAV flight plans. For spatiotemporal grids binding multiple UAV flight plans simultaneously, they are marked with varying degrees of busy, congested, or idle status based on flight business. Third, based on the flight plans associated with currently flying UAVs, the planned flight trajectories of all flying UAVs in the airspace are acquired. Combining the flight trajectory data of currently flying UAVs and their associated flight plan data, the start time range of each UAV's flight and the flight time of each UAV within each spatiotemporal grid are analyzed and determined. Based on the above data, all associated spatiotemporal grids are identified, and spatiotemporal grids available for flight business are further confirmed. Finally, based on the defined data range, select drone flight plans and flight spatiotemporal grid data that meet the range criteria; drone flight plan data includes drone type, planned flight airspace / flight route, planned flight time, flight mission nature and plan approval status; flight spatiotemporal grid data includes spatial coordinates, spatial size, spatial business attributes and flight time set data.

[0055] The UAV real-time operation data acquisition unit is used to acquire real-time flight trajectory data, flight spatiotemporal grid data, and real-time meteorological data of the UAV. Data fusion algorithms are used to filter and fuse these data to form flight data features, and the data is then standardized. This provides real-time operation data support for subsequent UAV operation risk prediction.

[0056] In this embodiment, real-time flight trajectory data of the UAV is obtained by connecting to the system data interface of the UAV operator or UAV manufacturer, UAV positioning terminal, UAV detection equipment, etc., and includes, but is not limited to: longitude, latitude, altitude, time, horizontal speed, vertical speed, pitch axis angle, roll axis angle, yaw axis angle, and remaining battery power. Real-time meteorological data is obtained from the flight spatiotemporal grid through the meteorological system data interface at the cubic meter level, and combined with the spatiotemporal grid to form new spatiotemporal meteorological data, including but not limited to: temperature, humidity, air pressure, wind speed, and visibility. For multiple data sources for the same UAV, this unit will analyze multiple data points, filter and fuse them using a data fusion algorithm to form the most accurate flight data features, standardize the flight data features according to the system's data format requirements, and associate the currently flying UAV with its flight plan through these flight data features.

[0057] The UAV operation risk assessment module combines flight trajectory prediction, collision grid detection, conflict space analysis, spatiotemporal grid flight time analysis, and real-time flight conflict event analysis during UAV flight to output future flight conflict events and optimal solutions. Compared to traditional risk assessment and analysis methods based solely on flight trajectory information, this system not only integrates current flight spatiotemporal environment information and flight trajectory information but also incorporates future flight spatiotemporal environment information and real-time changes in UAV flight trajectory information, achieving more accurate, efficient, and intelligent flight risk assessment and analysis.

[0058] The drone operation risk assessment module includes:

[0059] Flight trajectory prediction unit: Used to predict the future flight trajectory of the UAV using multi-dimensional attribute airspace grids, flight spatiotemporal grid data, and flight data features. It comprehensively considers parameters such as the UAV's current position, speed, acceleration, battery level, disturbance parameters, and meteorological information to output the future flight trajectory.

[0060] In this embodiment, refer to Figure 2 The prediction of the future flight trajectory of the drone is as follows:

[0061] S11. Clean and extract features from historical flight data, and combine multi-dimensional attribute airspace raster, flight spatiotemporal raster data and flight data features to establish a Gaussian mixture prediction model for UAV flight trajectory prediction.

[0062] S12. Define the disturbance of the UAV itself and the disturbance of the airspace environment, and combine them with the Gaussian mixture prediction model to output flight trajectory prediction data.

[0063] S13. Continuously compare the predicted flight trajectory data with the actual flight trajectory data to evaluate and adjust the Gaussian mixture prediction model.

[0064] The future flight trajectory of the drone is predicted based on the Gaussian mixture model. By increasing the drone's own disturbance and the environmental disturbance, the flight simulation is carried out. Its advantage is that the predicted flight trajectory takes into account both the drone's own and the environmental factors, and will be closer to the actual flight trajectory.

[0065] Collision Grid Detection Unit: Used to detect whether there are intersecting or close spatiotemporal grids in the future flight trajectory of the UAV. For spatiotemporal grids that intersect or are close to the future flight trajectory of the UAV, spatiotemporal conflict analysis is performed in combination with flight business attributes to determine whether there are spatiotemporal grids that may collide with the future flight trajectory of the UAV.

[0066] Conflict Space Analysis Unit: Used to perform spatial conflict analysis on spatiotemporal grids that have collisions.

[0067] Specifically, the spatial conflict analysis of the spatiotemporal grids where the UAVs are in collision involves analyzing and recording potential conflict points in the three-dimensional space horizontal and vertical directions of the spatiotemporal grids in collision using a spatial topology analysis algorithm. This involves combining geographical location coordinates, altitude information, and three-dimensional oblique image geographic information into a three-dimensional space for analysis.

[0068] Conflict Time Analysis Unit: Used to divide the conflict time of spatiotemporal grids where collisions occur. By combining flight plan data, actual flight data, and spatiotemporal grid data, it finely divides the different times when the UAV is stationary in the spatiotemporal grid, providing a time-dimensional analysis basis for flight conflict events.

[0069] Real-time Flight Conflict Event Analysis Unit: Used to determine future flight conflict events and find the optimal solution for such events.

[0070] Reference Figure 3 The determination of the future flight conflict events is as follows:

[0071] S21. Based on the current flight status of the UAV, combined with flight plan data, real-time flight data, and airspace grid data, the reachability set analysis algorithm is used to determine whether there is a conflict between the UAV and its relative position at a certain future moment.

[0072] S22. By continuously receiving actual flight data and real-time airspace environment information, the judgment result of step S21 is compared with the actual flight data, and the reachability set analysis algorithm is adjusted in real time to analyze the set of future flight trajectories that the UAV may reach within a certain time range.

[0073] S23. Analyze all points in the future flight trajectory set using the A* algorithm, eliminate data that is impossible to reach in the future flight trajectory set, and update the existing reachable dataset accordingly to determine future flight conflict events. All points include flight spatiotemporal raster data and flight trajectory prediction data.

[0074] The advantage of this real-time flight conflict event analysis method lies in the fact that the analysis process not only considers the flight dynamic characteristics of UAVs and the uncertainties of the flight spatiotemporal environment, but also fully utilizes the trajectory prediction capability of the reachability set algorithm and the efficient path search capability unique to the A* algorithm based on the fusion of multiple airspace information. This makes the predicted flight trajectory not only closer to the actual flight trajectory, but also outputs predicted trajectory data more efficiently and stably. This provides reliable, real-time and accurate flight trajectory data for real-time analysis of flight conflict events, enabling progressive analysis and correction to determine future flight conflict events.

[0075] Reference Figure 4 The optimal solution to the analyzed flight conflict events is sought, specifically as follows:

[0076] S31. Using flight conflict events as initial information, initialize the spatiotemporal grid, obstacles, and flight data of the conflict as information for each path.

[0077] S32. Summarize historical flight trajectory data and current airspace flight data, and decompose conflicting paths into multiple flight segments according to probability rules.

[0078] S33. Using the ant colony algorithm, find the optimal path that avoids conflict by consisting of multiple flight segments based on the pheromones on each conflicting path.

[0079] S34. Combining the current flight speed of the UAV and airspace environment information, assign time attributes to the starting points of the flight segments on each optimal path that can avoid conflict and the path that conflicts with the original analysis.

[0080] S35. An annealing simulation algorithm is adopted, with the optimal path, flight time, spatiotemporal grid and real-time flight information as data inputs, and airspace conflict events as the stopping condition. The algorithm finds the globally optimal solution in terms of time and path through repeated iterations.

[0081] In this embodiment, for the analyzed flight conflict events, the system follows the principle of minimum robustness, dividing the spatiotemporal grid where the conflict point is located into three segments: before, during, and after. Dynamic planning is performed on the flight path of the UAV in flight and the takeoff time or route of the UAV not in flight. The optimal solution for resolving flight conflicts is found by combining the Ant Colony Optimization (ACO) algorithm and the Simulated Annealing (SANA) algorithm. Compared to using only one algorithm to solve flight conflict events, the advantage of combining the two algorithms is that the Ant Colony Optimization algorithm provides a global optimization perspective for UAV flight in the airspace and finds multiple feasible flight paths to choose from, while the SANA iterative algorithm repeatedly finds the globally optimal solution among multiple solutions. This not only improves the efficiency and reliability of flight conflict analysis but also provides UAVs with flexible, diverse, and reliable flight conflict solutions. These solutions include, but are not limited to, adjusting the UAV's flight time, route, or other flight parameters to avoid flight conflicts and collision risks, ensuring the safe and efficient completion of UAV flights.

[0082] Drone Operation Risk Alarm Module: Based on the flight conflict events output by the system in real time, combined with different monitoring scenarios and risk configurations of the system, it provides drone risk alarm services and drone risk solutions.

[0083] The drone operation risk alarm module includes:

[0084] Risk Alarm Unit: This unit provides drone risk alarm services. When the risk value exceeds a threshold, the system immediately sends an alarm. This unit determines whether the drone poses a collision risk in four-dimensional space and issues an alarm. The alarm message issuance is based on risk analysis data. When the risk value exceeds the threshold, the system immediately sends an alarm to allow for necessary adjustments or emergency responses. The risk threshold can be configured with rules to adapt to different monitoring scenarios. Alarms can be issued via system pop-ups, icon markers, SMS alerts, email alerts, and WeChat messages to ensure timely and accurate delivery of alarm information to relevant personnel.

[0085] Risk Resolution Unit: This unit provides drone risk solutions, instructing the drone to adjust at least one of the following: flight altitude, flight speed, and flight path. It combines a rule engine and machine learning models to recommend solutions after a drone experiences an operational risk. Based on historical data and machine learning models, combined with risk assessment values, it intelligently recommends suitable risk solutions and provides multiple solution options, allowing users to choose the most suitable one based on the actual situation. All alarm events are recorded, including alarm content, time, and selected solution; these records can be used for future analysis and improvement. The unit learns from handling records and new data, continuously optimizing the recommendation algorithm and combining expert knowledge and the rule engine to improve the accuracy and reliability of recommendations.

[0086] The UAV simulation module is used to simulate flight plans and potential flight conflicts, and to simulate different UAV flight scenarios.

[0087] The UAV simulation module includes:

[0088] The simulation unit combines data from the UAV operation risk assessment module and the UAV operation risk alarm module to simulate flight scenarios under different flight conditions. It generates simulated flight data based on the system's functional data, mimicking flight data under various conditions and analyzing the development process of conflict flights. Simultaneously, it simulates and evaluates the consequences of conflicts of varying severity, generating visualized conflict analysis results. For different simulated conflict analysis results, it uses the aforementioned optimization algorithms to find optimal solutions, including adjusting flight altitude, speed, and flight path, to avoid or reduce the probability of flight conflict.

[0089] Visual Interface Unit: Provides users with a visual operating interface to display the results and solutions of drone risk assessment and analysis. It offers a user-friendly visual interface, showcasing the results of drone operation risk analysis and solutions, facilitating user decision-making and adjustments to flight plans. Simultaneously, it provides a real-time monitoring page and drone control interface, allowing for real-time monitoring of flight status during drone flight, dynamically visualizing conflict analysis results and solutions, and enabling users to promptly adjust drone flight plans to respond to emergencies.

[0090] As a preferred embodiment, after implementing this system in the airspace control system of an oil refinery, it demonstrated significant advantages over the previously used traditional conflict detection algorithms in inspection and security operations. It resolved practical flight operation problems such as untimely conflict detection, lack of suitable solutions when conflicts occurred, and inability to perform real-time dynamic conflict detection. Furthermore, the risk of flight operation conflicts was reduced by more than half, and the accuracy of flight conflict location and the early warning time for flight conflicts were significantly improved, greatly increasing the application rate of the effective solution. The following is a detailed explanation of the advantages and problems solved in the application scenario:

[0091] This system considers three-dimensional space (X, Y, Z) and time (T), and also comprehensively takes into account dynamic factors in the current airspace environment, such as movable obstacles, weather, the nature of UAV operations, flight speed, and predicted flight trajectories. It can accurately calculate the relative positions and movement trajectories of aircraft in all dimensions, and can predict the overall risk environment of the airspace where the UAV is located for early detection and avoidance of potential conflicts. In inspection and security operations, UAVs need to cruise at different altitudes and in complex terrain. Furthermore, the flight paths for security operations may change due to the impact of events. This system can more accurately predict flight paths, calculate conflict risks in real time, and avoid potential collisions with other aircraft or buildings, significantly improving the safety and reliability of operations.

[0092] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A system for detecting, analyzing, and resolving unmanned aerial vehicle (UAV) flight conflicts, characterized in that, The system includes: Data acquisition and preprocessing module: used to acquire and preprocess low-altitude basic environmental data, flight plan data, and real-time UAV operation data; The data acquisition and preprocessing module includes: Low-altitude basic environmental data acquisition unit: used to acquire geographic environmental data and airspace data to form air-ground information fusion data, and combine geospatial grid technology, system flight operation data, historical airspace flight data and future predicted airspace flight data to output multi-dimensional airspace raster; Flight plan data acquisition unit: used to acquire UAV flight plan data, analyze and integrate the time, space and low-altitude flight business attributes of UAV flight plan data, and perform rasterization processing, that is, to perform spatiotemporal raster matching and binding on the flight routes or flight areas within the UAV flight plan to determine the associated flight spatiotemporal raster data. The UAV real-time operation data acquisition unit is used to acquire real-time flight trajectory data and flight spatiotemporal grid data of the UAV, as well as real-time meteorological data. The data is filtered and fused through data fusion algorithms to form flight data features and is then standardized. Drone Operation Risk Assessment Module: This module is used to analyze drone flight trajectory prediction, collision grid detection, conflict space analysis, spatiotemporal grid flight time analysis, and real-time flight conflict event analysis during drone flight to output future flight conflict events and the optimal solutions to these events. The drone operation risk assessment module includes: Flight trajectory prediction unit: used to predict the future flight trajectory of UAVs using multi-dimensional attribute airspace raster, flight spatiotemporal raster data and flight data features; Collision grid detection unit: used to detect whether there are intersecting or close spatiotemporal grids in the future flight trajectory of the UAV. For spatiotemporal grids that intersect or are close to the future flight trajectory of the UAV, spatiotemporal conflict analysis is performed in combination with flight business attributes to determine whether there are spatiotemporal grids that may collide with the future flight trajectory of the UAV. Conflict Space Analysis Unit: Used to perform spatial conflict analysis on spatiotemporal grids that have collisions; Conflict Time Analysis Unit: Used to divide the spatiotemporal grids with collisions into conflict times; Real-time Flight Conflict Event Analysis Unit: Used to determine future flight conflict events and find the optimal solution for such events; The specific details of determining the future flight conflict events are as follows: S21. Based on the current flight status of the UAV, combined with flight plan data, real-time flight data and airspace grid data, the reachability set analysis algorithm is used to determine whether there is a conflict between the UAV and its relative position at a certain future moment. S22. By continuously receiving real-time actual flight data and real-time airspace environment information, compare the judgment result of step S21 with the actual flight data, adjust the reachability set analysis algorithm in real time, and analyze the set of future flight trajectories that the UAV may reach within a certain time range. S23. Use the A* algorithm to analyze all points in the future flight trajectory set, eliminate data that cannot be reached by the future flight trajectory set, and update the existing reachable dataset to determine future flight conflict events. The optimal solution to the analyzed flight conflict events is sought, specifically as follows: S31. Using flight conflict events as initial information, initialize the spatiotemporal grid, obstacles, and flight data of the conflict as information for each path; S32. Summarize historical flight trajectory data and current airspace flight data, and decompose conflicting paths into multiple flight segments according to probability rules; S33. Using the ant colony algorithm, find the optimal path that avoids conflict by combining multiple flight segments based on the pheromones on each conflict path. S34. Combining the current flight speed of the UAV and airspace environment information, assign time attributes to the starting point of the flight segment on each optimal path that can avoid conflict and the path that conflicts with the original analysis. S35. An annealing simulation algorithm is adopted, with the optimal path, flight time, spatiotemporal grid and real-time flight information as data inputs, and airspace conflict events as the stopping condition. The global optimal solution in terms of time and path is found through repeated iterations. Drone operation risk alarm module: It is used to provide drone risk alarm services and drone risk solutions based on the flight conflict events output by the system in real time, combined with different monitoring scenarios and risk configurations of the system; The UAV simulation module is used to simulate flight plans and potential flight conflicts, and to simulate different UAV flight scenarios.

2. The system for detecting, analyzing, and resolving UAV flight conflicts according to claim 1, characterized in that, Predicting the future flight trajectory of drones, specifically: S11. Clean and extract features from historical flight data, and combine multi-dimensional attribute airspace raster, flight spatiotemporal raster data and flight data features to establish a Gaussian mixture prediction model for UAV flight trajectory prediction. S12. Define the disturbance of the UAV itself and the disturbance of the airspace environment, and combine them with the Gaussian mixture prediction model to output flight trajectory prediction data. S13. Continuously compare the predicted flight trajectory data with the actual flight trajectory data to evaluate and adjust the Gaussian mixture prediction model.

3. The system for detecting, analyzing, and resolving UAV flight conflicts according to claim 1, characterized in that, Specifically, the spatial conflict analysis of the spatiotemporal grids where the UAVs are in collision involves analyzing and recording potential conflict points in the three-dimensional space horizontal and vertical directions of the spatiotemporal grids in collision using a spatial topology analysis algorithm. This involves combining geographical location coordinates, altitude information, and three-dimensional oblique image geographic information into a three-dimensional space for analysis.

4. The system for detecting, analyzing, and resolving UAV flight conflicts according to claim 1, characterized in that, The drone operation risk alarm module includes: Risk alarm unit: Used to provide drone risk alarm service. When the risk value exceeds the threshold, the system immediately sends an alarm message. Risk resolution unit: used to provide drone risk solutions, which are used to instruct the drone to adjust at least one of flight altitude, flight speed and flight path.

5. The system for detecting, analyzing, and resolving UAV flight conflicts according to claim 1, characterized in that, The UAV simulation module includes: Simulation unit: Used to combine data from the UAV operation risk assessment module and the UAV operation risk alarm module to simulate flight scenarios under different flight conditions; Visual Interface Unit: Used to provide users with a visual operating interface to display the results and solutions of drone risk assessment and analysis.

Images